analiza aiqbench
This research advances safe learning-based control by introducing uncertainty tubes derived from sampled realizations rather than restrictive smoothness assumptions, enabling robust handling of discontinuities in safety-critical dynamics. By integrating these tubes into a Bayesian optimization framework, the authors demonstrate practical parameter tuning on a Furuta pendulum, bridging theoretical rigor with experimental validation for complex robotic systems.
Safe learning-based control via function-based uncertainty quantification
arXiv:2604.01173v2 Announce Type: replace-cross Abstract: Uncertainty quantification is essential when deploying learning-based control methods in safety-critical systems. This is commonly realized by constructing uncertainty tubes that enclose the unknown function of interest, e.g., the reward and constraint functions or the underlying dynamics model, with high probability. However, existing approaches for uncertainty quantification typically rely on restrictive assumptions that encode smoothness properties of the unknown function, such as a known norm in a function space. Moreover, these methods usually struggle with discontinuities. In this paper, we model the unknown function as a random function from which independent and identically distributed realizations can be generated. We then construct uncertainty tubes via the scenario approach that hold with high probability. Our uncertainty tubes rely solely on sampled realizations and can therefore accommodate discontinuities represented by the sampling model. We integrate these uncertainty tubes into a safe Bayesian optimization algorithm with which we safely tune control parameters on a real Furuta pendulum.
Abstract:Uncertainty quantification is essential when deploying learning-based control methods in safety-critical systems. This is commonly realized by constructing uncertainty tubes that enclose the unknown function of interest, e.g., the reward and constraint functions or the underlying dynamics model, with high probability. However, existing approaches for uncertainty quantification typically rely on restrictive assumptions that encode smoothness properties of the unknown function, such as a known norm in a function space. Moreover, these methods usually struggle with discontinuities. In this paper, we model the unknown function as a random function from which independent and identically distributed realizations can be generated. We then construct uncertainty tubes via the scenario approach that hold with high probability. Our uncertainty tubes rely solely on sampled realizations and can therefore accommodate discontinuities represented by the sampling model. We integrate these uncertainty tubes into a safe Bayesian optimization algorithm with which we safely tune control parameters on a real Furuta pendulum.
Systems and Control (eess.SY); Machine Learning (cs.LG); Optimization and Control (math.OC)
arXiv:2604.01173 [eess.SY]
(or arXiv:2604.01173v2 [eess.SY] for this version)
https://doi.org/10.48550/arXiv.2604.01173 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 253 cuvinte
Subiecte conexe
Certifiable Safe Model-Based Reinforcement Learning with Control-Affine Dynamics Approximation
arXiv · 21 iul. 2026
A Risk-Sensitive and Uncertainty-Aware Decision-Making and Control Framework for Safe and Robust Autonomous Driving
arXiv · 10 sept. 2026
Worst-Case Distance-Aware Error Bounds for Neural Networks
arXiv · 7 aug. 2026
Do Uncertainty Signals Help? A Systematic Study of Uncertainty-Aware Decoding with Rollback Mechanisms
arXiv · 19 aug. 2026